Coding race
Every coding subcategory blended into one race: front end, back end, systems, debugging, agents and more.
- Leader
- Opus 5.5
- Margin
- 1.1 pts
- 1.2% ahead
- Weeks at #1
- 4
- in a row
- Changes
- 5
- in 2026
Warming up the track…
Re-weight this race
Drag to say what matters to you. The track re-sorts live and weeks-at-#1 are recomputed for your weighting.
Score history
Top 6 today over the leader's frontier · click to jump to a week
Reigns at #1
Longest: Opus 4.5, 11 weeks (from Nov 2025).
Inside this race
What makes up coding
Why this race matters
Coding is the single biggest professional use of language models and the clearest measure of real economic value.
How it is scored
Normalised subcategory scores, weighted toward agentic and debugging work.
Data provenance
Long-horizon software engineering tasks in real repositories.
Agentic tasks executed in a real terminal.
Blind votes on generated web apps.
Scores here are illustrative placeholders that mimic real benchmark patterns. Swap in live data through the score provider. Methodology